Regression Results for Marriage-Related Outcomes
| Divorced in Past Year (of Married/Recently Divorced) | Cohabiting (of Singles) | |
|---|---|---|
| Treatment*transition | 0.0025 (0.0016) | |
| Treatment*implementation | 0.0030* (0.0016) | |
| Treatment*enactment | −0.0093* (0.0048) | |
| Treatment*implementation year 1 | −0.0062* (0.0033) | |
| Treatment*implementation year 2+ | −0.0142*** (0.0021) | |
| Observations | 707,714 | 1,215,191 |
| Dependent variable means | ||
| Treatment, pre-enactment | 0.034 | 0.145 |
| Control, pre-enactment | 0.032 | 0.199 |
| Treatment, implementation | 0.035 | 0.130 |
| Control, implementation | 0.030 | 0.199 |
Source: 2008–2013 1-year ACS data. Notes: Shown are coefficients estimated from the specifications of the determinants of the probability of having divorced in the prior 12 months and the probability of cohabiting. Each column presents results from a single regression. Regressions are estimated using population weights. The divorce analysis sample includes individuals currently married and those divorced in the prior 12 months ages 23–25 and 28–30. The specification includes state, year, and age fixed effects, national and state-year trends, as well as individual demographic controls. The cohabitation analysis sample includes individuals currently unmarried ages 23–25 and 28–30. The specification includes state, calendar month, and age fixed effects, national and state-calendar month trends, as well as individual demographic controls. Specifications also include controls for the age-specific state-year sex ratio and unemployment rate: prior year measures are used for past year outcomes (divorce), and current year measures are used for current year outcomes (cohabitation). Robust standard errors clustered at the state level in parentheses,
↵*** p < 0.01,
** p < 0.05,
↵* p < 0.1.